Walt said:
But the Tamron 2.8/28~105 really caught my eye...
My first research stop was at Gary's site, and his opinion that
this lens is so dismal wide open that it's hard to focus was
almost enough to change my mind, but I think that=92s mainly at
28mm, so not all that serious. Gary's test results overall were
generally pretty good.
But then I found this:
http://www.connect.ab.ca/~buddy/lenst1.htm
------------------------------------------------
Well, I had a look and I think the lens tests leave a great deal to be
desired. In looking for the conditions under which the tests were
performed I found the following:
-------------------------------------------------------
"The target that I used was sized to the proportions of 24x36. Three
samples of 50x65 pixels were extracted from the full image which I
scanned at 2400 ppi, giving me an image size of approximately 2200x2400
pixels. They were taken from the centre, a near edge and a corner. This
gives a scanned image of approximately 23MB or approximately 7.48 mega
pixels.
A target was used of such a size as to show differences in resolution, a
target with minimal enlargement and all lenses would look great, with
too much enlargement of the field even the best lens would look bad so
the selected target size was a reasonable compromise.
There are some anomalies in the samples where the film plane was not
exactly parallel to the target, giving an image with a taper towards the
side where the two planes are furthest apart.
Some "waviness", especially in the later tests and using wider lenses
was caused by a slight buckling of the paper on the target. (Especially
after the target was dropped in the snow a couple of times.) :o)
The sloped lines of text that show in some samples, particularly the
corner sample, is caused by lens distortion.
The colours of the samples should be ignored as they were taken
outdoors, in snow and with sunny/overcast conditions."
-------------------------------------------------------
One comment immediately caught my attention. It says that the Tamron @
28mm was unable to focus close enough to fill the frame. Checking
further one can see that the target is actually a 7 column section of a
newspaper page. These lens tests are being performed at very close
range much as one would test a copy lens. I would estimate the
magnification ration of these images at about 1:9. Gary's tests are all
performed at 1:40. At such close range the flatness of field (or lack
therof) could be a significant factor since these are not copy lenses.
Rather than being poorly resolved the edge of the field may simply be
out of focus. There is no mention of focus other than letting the
camera do its auto focus thing on the central portion of the image.
Worse, there is comment that the film plane wasn't necessarily parallel
with the target all the time, the lighting conditions varied from
sunlight to overcast and the target may have been buckled from having
been dropped in the snow. I'm glad the tester makes no pretense to
scientific method because there sure isn't any here.
Now, how good is the Tamron 2.8/28-105? I don't know but listen to Gary
because you sure can't trust any result reported at this other site.
Chuck Norcutt
Woburn, Massachusetts, USA
< This message was delivered via the Olympus Mailing List >
< For questions, mailto:owner-olympus@xxxxxxxxxxxxxxx >
< Web Page: http://Zuiko.sls.bc.ca/swright/olympuslist.html >
|